Résumé
BACKGROUND: The prognosis of paraquat intoxication patients is poor and this makes the prediction of mortality important in administering aggressive treatment and admission. This article investigates the usefulness of simplified acute physiology score II (SAPS II), as a predictor of the mortality in paraquat intoxication. METHODS: We retrospectively reviewed 65 patients who were admitted in one hospital between January in 2005 and December in 2010. We calculated their SAPS II, serum paraquat level, and severity index of paraquat poisoning (SIPP) at the time of intensive care unit (ICU) admission. We investigated the relationship between each systems and the mortality. RESULTS: Overall mortality was 73.8%: 48 out of 65 patients died. Non-survived group (n = 48) had a higher SAPS II score (30.44 +/- 15.99) than survived group (n = 17 [15.7 +/- 6.26], p < 0.001). Serum paraquat level and SIPP were significantly higher in non-survived group than in survived group (p < 0.05, in all comparisons). By using the area under receiver operating characteristic curves (AUC), the SAPS II system yielded equal discriminative power (AUC = 0.82) with serum paraquat level (AUC = 0.896) and SIPP (AUC = 0.865). Hosmer-Lemeshow goodness-of-fit test C indicated SAPS II score validated well in paraquat intoxication group (p = 0.33). CONCLUSIONS: Serum paraquat level is the best way for prediction of mortality in patients with acute paraquat intoxication. If checking serum paraquat level is impossible or delayed, SAPS II score can be an alternative tool for evaluating the prognosis in paraquat intoxication.